Evaluation of a satellite-derived latent heat flux product in the South China Sea: A comparison with moored buoy data and various products

Article history: Received 28 April 2008 Received in revised form 2 October 2008 Accepted 3 December 2008 Latent heatflux in the South China Sea (SCS)mainly based on the Tropical RainMeasuringMission Microwave Imager (TMI) is comparedwith the Objectively Analyzed air-sea heat Fluxes (OAFlux), theGoddard Satellite-BasedSurface Turbulent Fluxes, version2 (GSSTF2), the JapaneseOcean Flux Data Sets with use of Remote Sensing Observations (J-OFURO), the European Centre for MediumRange Weather Forecasts (ECMWF), the National Centers for Environmental Prediction/ Department of Energy reanalysis 2 (NCEP2), and International Comprehensive OceanAtmosphere Data Set data (ICOADS). These products did not all compare well to in situ data. It is found that TMI latent heatflux andOAFlux approximately reproduce the time variations of buoy data. However, ECMWFand ICOADSwere found that hadmoderate systematic errors with respect to in situ data. All products qualitatively reveal a similar pattern in climatological fields. Although the satellite-derived products are quite similar, TMI latent heat is smaller than those of OAFlux, GSSTF2 and J-OFURO in the northern SCS. The difference between TMI latent heat and ICOADS is large, and the temporal correlation is extremely low in theentire basin except the LusonStrait. This is owned to that the ICOADShardly reproduces accurate variability in the data-sparse regions. Also the temporal cross-correlation coefficients of between TMI latent heat and ECMWF or NCEP2 are considerably lower in the central SCS, may be affected by the lack of ship observations of ECMWF and NCEP2 there. The present study also compares meridional structures of the zonal averaged latent heat. The TMI latent heat and OAFlux overestimate the other two satellite-derived products during summer, whereas in winter, the TMI latent heat underestimates the other products in the northern regions. Though the examination of the influence of input variables, we note that the underestimation of TMI latent heatflux than other three satellite derived products in the northern region is mainly caused by larger near-surface air humidity there. While, large sea surface saturation humidity contributes to larger ECMWF latent heat flux. On the other hand, lager latent heat flux of NCEP2 in central parts and ICOADS in eastern bands is possibly mainly caused by the difference in winds. © 2008 Elsevier B.V. All rights reserved.